DESCRIPTION: (Applicant's Abstract) Marijuana and alcohol are drugs that contribute to automobile accidents when used separately and in combination. A further danger to adults who drive after late-night drug use is sleep deprivation. While many post-accident blood analyses have implicated marijuana and alcohol, no controlled empirical studies have quantified the combined behavioral and subjective effects of sleep deprivation with either marijuana or marijuana-alcohol combinations. The novelty of this project is its use of controlled Sleep Laboratory measurements, a driving simulator, computerized body sway measurement, a test of divided attention, and self-report mood scales to assess the various ways sleep deprivation and alcohol impair a driver who uses marijuana. The project has three specific aims. The first aim is to quantify how alcohol alters the effects of marijuana on mood, equilibrium, attention, and simulated automobile driving. The second aim is to quantify how partial sleep deprivation alters the effects of marijuana on these measures. The third aim is to quantify the effects of sleep deprivation on marijuana-alcohol combinations. Three studies using randomized placebo controlled designs are proposed. In each study, subjects will smoke a marijuana cigarette (0, 1.77, or 3.95% delta 9 THC) in two or three separate conditions. Study 1 will manipulate a concurrent dose of alcohol (0.0, 0.5, or 0.8 g/kg). Study 2 will manipulate pre-session sleep hours (8 vs. 3). Study 3 will add the expectation and experience of alcohol to the conditions of Study 2, with each subject drinking a dose of alcohol (0.0, 0.5, or 0.8 g/kg) before testing. The dependent variables to be studied include 1) body sway as measured by computerized dynamic posturography, 2) subjective effects scales (Profile of Mood States and Visual Analog Scales), 3) the Stroop Color Test of divided attention, and 4) brake latency within the driving simulator. By clarifying the separate and combined effects of marijuana, alcohol, and sleep deprivation on these ecologically valid measures, the results will contribute to the prevention of drug-related highway fatalities.